viafb: fix LCD hardware cursor regression

Although I'd consider this a hardware bug, as there is hardware out that
for whatever reason does not support hardware cursors on LCD output we
have to care about it in the driver.  This fixes a regression (invisible
cursor) introduced by:

    viafb: cleanup viafb_cursor

Signed-off-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de>
Reported-by: Julian Wollrath <jwollrath@web.de>
Tested-by: Julian Wollrath <jwollrath@web.de>
Cc: Scott Fang <ScottFang@viatech.com.cn>
Cc: Joseph Chan <JosephChan@via.com.tw>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

diff --git a/drivers/video/via/viafbdev.c b/drivers/video/via/viafbdev.c
index d8df17a7d5fce50a0f2566f5ee06713eb0ae96bd..a0004c17af6c2133082289ced5d5fa9a4502bdbf 100644 (file)
--- a/drivers/video/via/viafbdev.c
+++ b/drivers/video/via/viafbdev.c
@@ -872,7 +872,9 @@ static int viafb_cursor(struct fb_info *info, struct fb_cursor *cursor)
        if (info->flags & FBINFO_HWACCEL_DISABLED || info != viafbinfo)
                return -ENODEV;
 
-       if (chip_name == UNICHROME_CLE266 && viapar->iga_path == IGA2)
+       /* LCD ouput does not support hw cursors (at least on VN896) */
+       if ((chip_name == UNICHROME_CLE266 && viapar->iga_path == IGA2) ||
+               viafb_LCD_ON)
                return -ENODEV;
 
        viafb_show_hw_cursor(info, HW_Cursor_OFF);